A few weeks ago I participated in a panel on ethics at a ‘Future of Finance’ conference at Oxford’s Saïd Business School. An audience member asked our panel to comment on herding behaviour in banks, and since then I’ve been giving this issue some more thought. To what extent can human herding be explained in terms of the same goals that motivate herding in other social species?

In human societies, herding often involves people using the actions of others as a guide to sensible behavior, instead of independently seeking out high-quality information about the likely outcomes of these actions. Herding can be particularly destructive in market contexts, because blind faith in market trends by a swarm of individuals can lead to huge bubbles and devastating crashes. But if herding can lead to outcomes that are so damaging and maladaptive at the level of the society, then why did it evolve in the first place? Because herding evolved to benefit individuals, not groups or societies.

We’re used to thinking of social groups as fundamentally cooperative entities, but with some kinds of groups, nothing could be further from the truth. In fact, the best-known biological theory of herding, William Hamilton’s “selfish herd” idea [1], proposes that herds are the result of individuals trying to ensure that other members of their species, rather than themselves, will get eaten by predators. According to this theory, in many social aggregations, the risk of predation is higher at the periphery than at the center. A herd’s form and movement can be the result of individuals competing to stay close to this center, so that other individuals end up between themselves and the predators. Selfish herds have been proposed to occur in many different species, from wasps to guppies to sheep [2-4].

Read more at Psychology Today

Published On: July 7, 2013

Michael Price

Michael Price

Michael E. Price is Senior Lecturer in Psychology, and co-Director of the Centre for Culture and Evolutionary Psychology, at Brunel University London. He has a BA from Duke University and a PhD from the UC Santa Barbara Center for Evolutionary Psychology, and he has conducted studies among both Westerners and indigenous Amazonians. His research focuses mainly on the evolutionary origins of moral beliefs, especially those related to cooperation, punishment, egalitarianism, leadership, and sexual behavior.



  • Helga Vieirch says:

    Human children are profoundly influenced by the stories they are told and the exemplars they are shown. These stories and exemplars (what Dennett suggests we consider the more powerful of cultural memes) all all of to work on human nature illustrating what behaviours will lead to admiration and acceptance and which will lead to ridicule or punishment. This generates a certain conformity within groups, when reacting to events to which they have been conditioned to assume certain kinds of individual cooperation are expected.

    Humans live with social groups, however, these cannot really be compared to herds, or even to troops or packs. For one thing, most humans, whether they are living in a hunter-gatherer camp or a large city, are not always in physical proximity to the same set of relatives or friends. They do not move around the landscape in these groups. They tend to move around the landscape independent of their immediate residential group. A hunter may spend most of his day visiting a friend in another band (camping party) or out hunting by himself or checking a trapline. A woman might spend her day walking out with a few of her campmates to gather wild plant foods, or she might spend her day in camp keeping an eye on everyone’s children and making sure nobody’s hearth fire goes out….and watchful for possible predators the whole time.

    Family residential groups in hunter-gatherers might cluster together for ease of sharing and cooperation in bands. But bands regularly split up and move location, as families move out due to conflict or to previous arrangements to go camp with somebody else in a different location; and the campsite of the band is often abandoned within a matter of weeks. Bands are not herds.. they rarely even move as a group through the landscape. Occasionally a few families will travel to a new camping spot together, but it is not herding instinct that drives this. It is an optional and opportunistic, not an obligatory behaviour. When people file into the same bank or restaurant this is not usually as “groups” – they arrive as individuals most of the time and follow conventions of “polite” patron behaviour typical of their culture. Setting up rope-guides to “herd” people waiting in line is not really comparable to what happens when you watch a school of fish swimming.

    None of the social groups we can think of in humans involve “herd behaviour” that can be comparable to that of sheep or wildebeest or elephants. Indeed, even the idea that the grouping or herding behaviour of sheep is motivated by selfish concerns is overstating the case. Certain individuals benefit from being within a social group in that there are more eyes to spot a predator, and to give everyone else the warning to flee. Certainly predators chasing herd animals will pick out the sick, the injured, and the young individuals. But this doe snot mean that survival chances of those – or there fitter individuals, would necessarily have been greater if they were not in the herd. Herd dynamics and flight behaviour does not always involve “selfish” behaviour with individuals trying to hide in the centre. Honestly, it is embarrassing to have to point out that extreme flocking behaviour in domestic sheep has been selcted for BY HUMANS so the animals could be more easilly controlled, and is not an unambiguous indication of selfish genes, of the operation of natural selction, nor even, of the behaviour of all breeds of sheep. Breeds like the Soay and the Icelandic tend to scatter when threatened, even by human herders, let alone by predators, as do wild species of sheep, notewithstanding the fact that these animals all tend to live in groups. So the extreme flock behaviour, where each animal tries to get into the centre of the flock, is not proof that group slection does not “work”. It is more likely the product of human selection, for a behaviour during panic in sheep, that actually counters the behaviour “nature” selected for, which was scattering, and which resulted in the predators having to zero in on one animal. Such a behaviour is clearly of benefit to the survivors, who are a) likely to be close relatives, and b) likely to benefit over the long run by being members of groups where individuals “sacrifice” themselves to permit the rest to flee.

    All sheep aside, when we get to humans, as Wilson’s new thesis does, (and with intent), group selection is often pictured, even by some misguided anthropologists, as a matter of small groups striving to outdo one another in holding on to the most land and by stealing each other’s females. The assumption of people who write about this (like Lawrence Keeley, Richard Wrangham, and Steve LeBlanc) is that early human groups were assembled around a bunch of closely related males. The code word for the kind of small scale warfare these authors invisage is cooperative coalitional violence, or words to that effect.

    What has seemly escaped their notice is that humans evolved as mobile foragers, and mobile foragers rarely have lineages (based on male descent or female descent!), let alone “groups” in the current way this term is understood, as meaning a relatively permanent assembly of individuals.

    Among most hunter-gatherers kinship reckoning is bilateral, there are no lineages, no clans, and there are rarely fixed villages with a stable population. Women are not “property” but economic partners of men, often responsible for over 75% of all food production, water-fetching, much of the house construction, and the provisioning of material comforts like firewood.

    There is no real social stratification, although there might be informal ranking of individuals and households based on their diligence, generosity, and interpersonal skills. Indeed, within a functional forager economy, the kind of permanent villages, reduced individual mobility, reduced fluidity of population distribution over the landscape, which such political entities as lineages and clans imply, are arrangements awkward if not impossible to sustain.

    This may seem a trivial point, until you recall that foraging is the original human economy. The one that defined social action and survival – and many of the selection pressures acting upon human beings until fairly recently. Recent kinds of kinship and political groupings may, for all we know, atypical of the species and contrary to the conditions that would have existed throughout much of human evolution. So models based on “coalitional” male groups fighting each other and stealing females from each other, resulting in “group selection” for the fiercest groups, are farfetched.

    Kin-selction models, even when they extend to fictive kin, do actually sort of hold up under testing. The extension of altruism beyond known kin is, however, characteristic of all human cultures, however, so they do not adequately explain all of the actual data.

    Models based on the extension of altruism beyond immediate kin, however, do not require any sort of group, or any kind of male (or female) coalitions of any permanent kind. In fact the extension of kinship and proxy-kinship networks of sharing and mutual aid, in all directions *from each individual* no matter where this individual might be living within a cluster of several hundred camps that constitute a local linguistic community, fits in very nicely with what we do know about how hunter-gatherer economies optimize resource use and long term sustainability.

    Even in some species of animals, like horses, there is a changeable number of individuals within any group, groups tend to be temporary and based as much on friendship as on kinship, and there is not even clearcut leadership. It would be very hard to speak of group selection in anything smaller than a regional collection of scattered bands – a bachelor bunch here, a couple of mares with a stallion here, a lone mare with a newborn foal looking for another herd to join trotting off over there…
    Who “herds” other animals? Usually, a predatory species that hunt cooperatively. Lions, wolves, and African hunting dogs come to mind. And humans. To avoid “being herded” many specie shave developed tactics to confuse the predators: springing high into the air like impala, or taking flight in a disorganized fashion with individuals frequently changing positions relative to one another… or just scattering in all directions, like wild sheep.

  • […] more people are likely to believe it. Combine this truism with the fact that human beings prefer to move in herds (i.e. “go with the flow”), and you have some pretty interesting societal norms that develop […]

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.